From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 11532 invoked by alias); 16 Mar 2003 14:56:01 -0000 Mailing-List: contact gcc-prs-help@gcc.gnu.org; run by ezmlm Precedence: bulk List-Archive: List-Post: List-Help: Sender: gcc-prs-owner@gcc.gnu.org Received: (qmail 11442 invoked by uid 71); 16 Mar 2003 14:56:01 -0000 Resent-Date: 16 Mar 2003 14:56:01 -0000 Resent-Message-ID: <20030316145601.11441.qmail@sources.redhat.com> Resent-From: gcc-gnats@gcc.gnu.org (GNATS Filer) Resent-Cc: gcc-prs@gcc.gnu.org, gcc-bugs@gcc.gnu.org Resent-Reply-To: gcc-gnats@gcc.gnu.org, a.cilio@et.tudelft.nl Received: (qmail 3834 invoked by uid 48); 16 Mar 2003 14:49:05 -0000 Message-Id: <20030316144905.3831.qmail@sources.redhat.com> Date: Sun, 16 Mar 2003 14:56:00 -0000 From: a.cilio@et.tudelft.nl Reply-To: a.cilio@et.tudelft.nl To: gcc-gnats@gcc.gnu.org X-Send-Pr-Version: gnatsweb-2.9.3 (1.1.1.1.2.31) Subject: other/10109: documentation of target macro LIBCALL_VALUE X-SW-Source: 2003-03/txt/msg01070.txt.bz2 List-Id: >Number: 10109 >Category: other >Synopsis: documentation of target macro LIBCALL_VALUE >Confidential: no >Severity: non-critical >Priority: low >Responsible: unassigned >State: open >Class: doc-bug >Submitter-Id: net >Arrival-Date: Sun Mar 16 14:56:01 UTC 2003 >Closed-Date: >Last-Modified: >Originator: Andrea Cilio >Release: any from 2.7 to 3.2 >Organization: >Environment: not relevant >Description: The description of target macro LIBCALL_VALUE ( ) in the internals manual states: "[...] The definition of LIBRARY_VALUE[sic] need not be concerned [with] aggregate data types, because none of the library functions returns such types." I assume that LIBCALL_VALUE is intended where LIBRARY_VALUE is mentioned (the latter does not exist). Another problem is that it is not clear whether this macro must take into account that some library support functions, although not returning aggregate types, may require to return their result in memory like aggregates and BLKmode values. This is the case of functions that return "double" and emulate double precision floating point on targets with only 32-bit integer registers. Or should "need not be concerned" be interpreted as the fact that gcc cannot handle support functions that return their result in memory? >How-To-Repeat: >Fix: >Release-Note: >Audit-Trail: >Unformatted: